There is continuing controversy concerning the potential role of dietary fiber in the prevention of colorectal cancer. The results of animals studies, human epidemiologic studies and mechanism of action evaluations suggest an inverse relationship between wheat bran fiber intake and large bowel cancer incidence. Recently, in a pilot study, investigators at the Arizona Cancer Center have reported that a supplement of 13.5 gm/day of wheat bran fiber for 8 weeks (after 4 weeks of a placebo run-in period) significantly decreased (p <.001) [3H]-thymidine labeling in crypt organ and 24-hour outgrowth cultures of rectal biopsied obtained from patients with surgically resected colorectal cancers. In October, 1987, a randomized Phase II trial (NCI #PO1 CA41108) was initiated in recently resected colorectal adenoma patients comparing wheat bran fiber (13.5), calcium (1500 mg) and the combination of wheat bran fiber plus calcium versus placebo (2.0 grams wheat bran fiber and 250 mg elemental calcium) in Sun City, Arizona. The endpoints for analysis of this ongoing trial include: (1) inhibition of [3H]-thymidine labeling in rectal mucosal crypts and 24-hour outgrowth cultures of unaffected flat mucosal biopsy samples, and (2) lowering of fecal bile acid concentrations, specifically that of deoxycholic acid. This Phase II randomized trial achieved full patient accrual in February 1989. Because 100 fully available patients were needed, it was necessary to recruit 5 additional patients by August 1, 1989. Completion and final analysis will be performed in June 1990. The proposed Phase III trial will be initiated in July 1990. Because there appears to be no evidence of fiber toxicity during the 9 month period of the ongoing Phase II trial, we plan to evaluate fiber (versus placebo) as an effective intervention in the prevention of adenoma recurrence in 950 patients in the proposed two-armed Phase III trial. Secondary analysis endpoints for the Phase III trial in a subset of 250 adenoma patients will include the evaluation of the effect of the intervention on [3H]-thymidine labeling in crypt organ and 24-hour outgrowth cultures in rectal mucosal biopsy samples. This document outlines the background, significance, aims and procedures of the Phase III adenomatous polyp recurrence trial involving wheat bran fiber as the agent of choice.